Search

Popularity of e-commerce is a result of increasing needs of the community due to the ongoing restrictions on activities during the COVID-19 pandemic. Therefore, the role of digital technology needs to be further developed and applied in various economic sectors to balance the needs of society with the availability of existing products. The large number of products with specifications, prices, and types offered by e-commerce platforms provides a unique complexity for users to determine the choice that suits their needs. Aquascape is one of the tertiary needs in the form of hobbies, segmented product needs and is in great demand by both individuals, groups and industries. The great interest in aquascape is due to the effect of the treatment produced through soothing and therapeutic visuals of an aesthetically pleasing artificial natural environment. The range of products and price determines the quality and overall value of the aquascape. This is one of the considerations in choosing the right product to build an aquascape set. In addition, competition in e-commerce can confuse potential buyers with competing products with price volatility. However, the relevance of the right product can be achieved through web crawling techniques. The application of this technology allows users to collect product information specifically and automatically based on data collected from several e-commerce websites. Applications with this technique are built in the form of a search engine system that allows users to enter certain keywords in the context of aquascape to get relevant specific related products. [ABSTRACT FROM AUTHOR]

Liver disorders in humans do not see age, all age groups from young to old can be affected by liver disorders. Liver disorders can be medically broken down into 3 more specific types, namely hepatitis C, fibrosis and cirrhosis. Hepatitis C is generally caused by a virus, fibrosis is caused by genetic factors and cirrhosis is caused by the emergence of tissue in the liver. To make a specific diagnosis of this type of liver disorder, it is necessary to conduct research related to prediction of human liver function disorders using machine learning methods. There are many machine learning methods, and on this occasion the researcher is also trying to find the philosophy of computer science from this research which will look for evidence of the truth of the performance of each machine learning method that makes predictions in this study. The end result is axiological philosophy of science from the analysis of the truth of the performance of this study which states that the performance of the neural network has a greater accuracy of 94.4% and is followed by other methods such as nave Bayes 81.3%, kNN 93.6%, and SVM 94.1%. [ABSTRACT FROM AUTHOR]

In the philosophy of science, humans are seen as thinking creatures and can use their minds. Computer science is part of science where knowledge is first tested for truth through the scientific method. Philosophy and technology have a close relationship. Philosophy is the basis of thinking that gives birth to science that will continue to develop and produce products in the form of technology. Then humans use technology to make it easier to achieve goals. Currently, the problem is also in manually identifying the types of medicinal plants. There are difficulties in distinguishing types of medicinal plants due to the similarity of shape and color in the background for developing knowledge by utilizing machine learning and computer vision. The philosophy of science became a foundation for the development of computer science. The result is carried out using three machine learning methods, namely decision tree, KNN, and neural network, which are philosophical concepts connected with human life principles. By carrying out a systematic and structured scientific method and good truth testing in accordance with the philosophical perspective, it is found that the accuracy of the neural network method is 94%, the accuracy of the KNN method is 77.9%, and the accuracy of the DT method is 65.1%. Based on the evaluation of recall performance, precision, and F1 score, it is known that the neural network method can be used to identify medicinal plants. [ABSTRACT FROM AUTHOR]

Sumbawa Island is part of the territory of Indonesia which is in the position of three active earth plates which of course will have an impact on frequent earthquakes and the risk of earthquake hazards. Research on earthquakes in general has been done a lot, but especially on the island of Sumbawa there are still few who do research on the dangers of earthquakes. Earthquake research has uncertain parameters and to obtain optimal results an appropriate method is needed. In general, some predictive data mining methods are grouped into two categories, namely Parametric and Non-Parametric. This study uses a non-parametric method with MARS. There are two stages of MARS completion, namely Forward Stepwise and Backward Stepwise algorithms. The results of this study after testing the parameters and analysis obtained a mathematical model with 11 basis functions (BF) that contribute to the response variable, namely the basis function (BF) 1,2,3,4,5,7,9, and 11. not contributing are BF 6.8, and 10. The predictor variables that have the greatest influence are 100 % Epicenter Distance and 73.8% Magnitude. It can be concluded that the results of the prediction analysis of the areas in Sumbawa that have the highest earthquake hazard are Mapin Kebak, Mapin Rea, Panjang Island and Saringi Island. [ABSTRACT FROM AUTHOR]

Due to the global situation, it is necessary for an industrial company to integrate production scheduling and traceability planning, in a way to get more reliable stand and to make economics advantages. This paper presents biomass supply chain company which could produce a renewable energy. Traceability is a tracking requirement to check whether the quality of the product is satisfied. The supply chain of biomass planning problem aims to meet customer demand subject to traceability and other restrictions. The problem is modelled as a mixed integer linear program, and then it is solved using neighborhood search approach. [ABSTRACT FROM AUTHOR]

The classic data envelopment analysis (DEA) concept assumes that decision making unit (DMU) inputs and outputs can be clearly measured because they have definite values. However, there are times when there is uncertainty in a situation. This study looks at how to build a DEA model with input and output uncertainty. In the DEA model, the uncertainty situation is approached using a stochastic concept. Furthermore, the robust optimization method, which can solve the uncertainty problem, is introduced. As a result, at the conclusion of this study, a robust optimization model based on the Stochastic DEA concept is introduced. [ABSTRACT FROM AUTHOR]

Data Envelopment Analysis (DEA) is a linear programming-based technique used to evaluate the relative efficiency of multi-input and multi-output decision-making units based on observed data by comparing one decision-making unit (DMU) with another DMU. The efficacy assessment process sometimes involves a stochastic approach due to the uncertainty inherent in many real life problems, so the DEA model discussed in this paper is the stochastic DEA (SDEA) model. Data-driven decision-making is based entirely on the data involved. This paper proposes a framework for the completion of SDEA based on a data-driven approach. [ABSTRACT FROM AUTHOR]

k-Nearest Neighbor (k-NN) has fairly good accuracy results on data with almost the same class (balanced), whereas for data whose class distribution is not balanced, the accuracy value of k-NN tends to be low, moreover k-NN does not differentiate each feature data, meaning that each feature has the same weight in determining the new data class so that before classifying the feature selection process it is necessary to select the features that are most relevant to the data class. In this research, to overcome this problem, we will propose a framework using the Synthetic Minority Oversampling Technique (SMOTE) method to solve the class imbalance problem and the Gain Ratio (GR) to perform feature selection, so that get a new dataset with a balanced class distribution and features relevant to class data. The datasets used in this research such as E-Coli, Glass, and New Thyroid. For objective results, the 10-fold-cross validation method will be used as an evaluation method with k values 1 to 10. The results of the research prove that SMOTE and GR can increase the accuracy of the k-NN method, where the highest increase occurred in the Glass dataset by a difference increase of 18.43%. The lowest increase in accuracy occurred in the new thyroid dataset with an increase of 8.73%. The increase in accuracy occurred in all datasets used with an average increase of 12.87 [ABSTRACT FROM AUTHOR]

Quantum computing is a technology that takes advantage of quantum effects, which operates using quantum computing, resulting in increased algorithm speed compared to the speed of algorithms that use classical computing. This study aims to classify bank marketing data in determining the eligibility of customers who are entitled to become depositors with the quantum perceptron algorithm using calculations with 5 variables, namely employment, status, education, home credit, and bank loans. The result of this research is the analysis of bank marketing classification data to determine the eligibility of customers who are entitled to become depositors using the quantum perceptron algorithm with quantum computation. [ABSTRACT FROM AUTHOR]

The simple Additive Weighting method is one of the methods of Decision Support System which is often referred to as weighted summation method. But the weight value on this method is not officially the calculation used. Therefore, usually, several researchers combine this method with other methods to be more precise and accurate in supporting the decision. In this study, the authors compared the results of the SAW method decisions between weighting based on the Fuzzy Logic method and weighting based on the Rank Order Centroid (ROC) method. The case studied was the amount of student satisfaction with learning outcomes during the Covid-19 pandemic. The result is the number of students who expressed satisfaction with learning during the Covid-19 pandemic as many as 26 students for weighting Fuzzy Logic method and 23 students for weighting Rank Order Centroid method (ROC). [ABSTRACT FROM AUTHOR]

To optimize flight planning is an essentialpoint for every airline company must consider such as maximizing profits. Several stages involved in flight planning optimization, for example, the scheduling of the aircraft fleet, which should be planned optimally how to schedule the aircraft to different multi-leg flights. Howeverin reality the demand is uncertain at this early stage due to a variety of factors. We formulate a mixed integer stochastic programming model for solving the planning problem. [ABSTRACT FROM AUTHOR]